Bioequivalence and Bioavailability Forum • Pocock alpha's

Pocock alpha's [Two-Stage / GS Designs]

posted by d_labes – Berlin, Germany, 2011-02-16 10:18 (5240 d 09:50 ago) – Posting: # 6631
Views: 26,726

Ahoy ElMaestro,

❝ However, I do not have an understanding of the way Potvin et al. derive their alpha values - ...

Here we are already two :-D

.

As far as I understood: The alpha's are taken from the literature about group-sequential designs, especially the Pocock boundaries. See this nice presentation for an illustration.
The boundaries and alphas were derived based on

test for difference
large sample tests, i.e. Z test statistics which are multivariate normal distributed
interim analysis half way to a pre-chosen Nmax

All these preconditions do not hold for BE testing with an interim sample size adaption! Nevertheless they used the alpha's and succeeded in showing that the overall alpha is bounded at 0.05.
In that sense the alphas are purely empirical, I think.
The real alphas you can take from your simulations. Count the accepted simulated studies at each stage for the simulations with GMR=1.25 or 0.80.

So far the opinion of an "interested amateur" (cit. HS) in statistics.

—
Regards,

Detlew

Complete thread:

Potvin method B subtleties d_labes 2011-01-31 14:24
- Potvin method B subtleties jdetlor 2011-02-01 17:17
  - Potvin method B subtleties d_labes 2011-02-02 09:15
    - Potvin method B subtleties jdetlor 2011-02-02 16:08
      - Gould d_labes 2011-02-02 16:25
        
        Gould jdetlor 2011-02-02 17:02
- Another one ... d_labes 2011-02-07 14:32
  - Oops! Helmut 2011-02-07 15:24
    - Oops! d_labes 2011-02-07 16:29
      - Oops! Helmut 2011-02-07 16:47
        
        Oops! ElMaestro 2011-02-07 17:06
        
        Oops! Helmut 2011-02-07 17:17
        
        Oops! ElMaestro 2011-02-07 18:35
        
        Simul'ants in my pants d_labes 2011-02-08 09:07
        
        Need for speed... ElMaestro 2011-02-13 10:57
        
        Need for speed... d_labes 2011-02-15 14:06
        
        Need for speed... ElMaestro 2011-02-15 17:39
        
        Pocock alpha'sd_labes 2011-02-16 09:18
        
        ElMaestro's brain ElMaestro 2011-02-16 09:32
        
        ElMaestro's brain d_labes 2011-02-16 10:05
        
        ElMaestro's brain ElMaestro 2011-02-16 12:07
        
        Numbers d_labes 2011-02-16 13:45
- Bonus counter-question ElMaestro 2011-02-07 23:44
  - Speed is the question d_labes 2011-02-08 08:36
- New paper (method D for PE 0.90) Helmut 2011-04-18 12:58
  - New paper (method D for PE 0.90) d_labes 2011-04-19 08:58
    - New paper (method D for PE 0.90) Helmut 2011-04-19 14:30
    - New paper (method D for PE 0.90) ElMaestro 2011-04-19 14:49
      - Stopping criterion # futility rule Helmut 2011-04-19 15:28